Combining High-Frequency Trading-Behavior Factors for Equity Strategies

Summary

This equity research summary describes high-frequency factors built from combinations of price and trading-volume information. It presents two examples: an illiquidity factor adjusted using changes in the price path, and a trading-behavior factor derived from aggressive buy and sell orders. The reported tests find that these factors have some relationship to established style and higher-moment factors, but retain distinct information. The note evaluates their performance after removing the linear influence of size for one factor and, in a separate portfolio exercise, removing Barra factor exposures.

The authors also discuss limits of high-frequency signals. Individual factors can contribute new information even when their returns overlap, and adding a factor may provide limited incremental returns; downside risk and tail exposures can remain correlated. They report that combining factors with equal weights improved the overall strategy’s measured stability and risk profile in their tests. These are historical results on the full A-share universe described in the summary, with no detailed information here about trading costs, data quality, or live performance, so they do not establish that the factors will generalize.

Key ideas

• Price and volume combinations can produce high-frequency factors with information distinct from traditional factors.
• The examples use price-path changes to adjust an illiquidity signal and aggressive orders to measure trading behavior.
• Factor information may be distinct even when factor returns overlap or provide limited incremental gains.
• Downside and tail risks can remain correlated across high-frequency factors.
• The note reports that an equal-weight combination improved measured strategy stability in historical tests.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.